[SOLVED] count

Micky12345 · 03-02-2012, 06:25 AM

i have written a code to find the number of alphabet 'a' in a file

my code is below

Code:

count=`cat $fname | grep -ic 'a' `
echo $count

where $fname is name of a file

my file contents are:

Aathira
apple
pineapple and
orange

Actually answer should be 7 but its coming as 5
In the string 'Aathira' only one 'a' is counted
How it can be rectified??

catkin · 03-02-2012, 06:33 AM

Code:

sed 's/[^a]//g' input.txt | tr -d '\n' | wc -c

coralfang · 03-02-2012, 06:45 AM

Firstly the answer should be 4, rather than 5.
The -c option displays how many lines contain an 'a', not the total count of the letter 'a'.

This seems to work:

Code:

count=`fold -w 1 $fname | grep -ic 'a'
echo $count

Explanation:
Print each character of the file's contents line by line, and then pass it to grep.

Cedrik · 03-02-2012, 07:09 AM

Quote:

Originally Posted by catkin

Code:

sed 's/[^a]//g' input.txt | tr -d '\n' | wc -c

Nice, but it seems case insensitive match was a requirement, so i modifier could be used

Code:

count=$(sed 's/[^a]//ig' $fname | tr -d '\n' | wc -c)

Another way

Code:

count=$(perl -ne '$i++ while /a/ig; END{print "$i\n"}' $fname)

whizje · 03-02-2012, 07:35 AM

another possibility

Code:

grep -oi 'a' test|wc -w